Working Paper
WP no 699
July, 2007
SUBSIDIARY STRATEGY:
THE EMBEDDEDNESS COMPONENT
Carlos García-Pont
J. Ignacio Canales
Fabrizio Noboa
IESE Business School – University of Navarra
Avda. Pearson, 21 – 08034 Barcelona, Spain. Tel.: (+34) 93 253 42 00 Fax: (+34) 93 253 43 43
Camino del Cerro del Águila, 3 (Ctra. de Castilla, km 5,180) – 28023 Madrid, Spain. Tel.: (+34) 91 357 08 09 Fax: (+34) 91 357 29 13
Copyright © 2007 IESE Business School.
IESE Business School-University of Navarra - 1
SUBSIDIARY STRATEGY:
THE EMBEDDEDNESS COMPONENT
Carlos García-Pont*
J. Ignacio Canales**
Fabrizio Noboa***
Abstract
This paper inductively derives a model that develops the concept of subsidiary embeddedness as
the framework within which subsidiary strategy can take place. Our model identifies three
hierarchical levels of embeddedness. Operational embeddedness relates to interlocking day-today relations. Capability embeddedness deals with the development of competitive capabilities
for the multinational as a whole. Strategic embeddedness deals with subsidiary participation in
the MNC strategy setting. We deem these three types of embeddedness as ways to develop
subsidiary strategic alternatives. As such, different types of subsidiary embeddedness imply
different subsidiary roles. Embeddedness, as we inductively concluded from a revealing case
study, is not merely an outcome of institutional setting, but a resource which a subsidiary can
manage by means of manipulating dependencies or exerting influence over the allocation of
critical resources. A subsidiary can modify its embeddedness to change its strategic restraints.
Therefore, the development of subsidiary embeddedness becomes an integral part of subsidiary
strategy.
*
Professor, Marketing, IESE
** Lecturer, University of St Andrews
*** Professor, USFQ Business School
Keywords: Multinational management, subsidiary, strategy, organization.
IESE Business School-University of Navarra
SUBSIDIARY STRATEGY:
THE EMBEDDEDNESS COMPONENT
Introduction
Theory and research have suggested that subsidiaries may develop strategy alongside their
evolution (Birkinshaw and Hood, 1998; Taggart, 1999; Andersson, Björkman and Forsgren,
2002). At the same time, embeddedness of any firm in its business network has significant
implications for its performance (Zukin and DiMaggio, 1990; Uzzi, 1996; Dyer and Singh, 1998;
Håkansson and Snehota, 1998; Gulati, Nohria and Zaheer, 2000; Rowley, Behrens and
Krackhardt, 2000; Hung, 2002). In fact, embeddedness is a good predictor of the role a
subsidiary may play in the overall Multinational Corporation (MNC) network (Andersson and
Forsgren, 1996) and the subsidiary's local embeddedness partially explains subsidiary
knowledge creation (Taggart and Hood, 1999; Andersson, Björkman and Forsgren, 2005).
Subsidiaries, however, have more constraints in developing their own strategy given that they
are embedded in different networks that include all other units of the multinational to which
they belong, alongside customers, suppliers and other institutions. Moreover, they typically
have a pre-set business domain that limits their possible strategy (Birkinshaw and Hood, 1998).
Inspite of subsidiaries face corporate and resource constraints, and the establishment of lateral
relations between MNC units (Birkinshaw and Morrison, 1995), they actually have space to
define their own mission and goals, and look after their own future (Birkinshaw and Hood,
1998), which influences performance (Subramaniam and Watson, 2006). Nevertheless, the
process whereby and the extent to which subsidiaries develop their own strategy via
embeddedness has been given less attention.
This paper uses the notion of embeddedness to inductively develop a model that contributes to
the literature on subsidiary strategy (Birkinshaw and Hood, 1998). The contribution of this
paper is to present a model that uses the very embeddedness of the subsidiary as the key driver
of subsidiary distinctiveness. We consider distinctiveness an integral part of subsidiary strategy,
as it builds the subsidiary’s raison d’être within the multinational network (Ghoshal and
Bartlett, 1990). Our purpose is to inductively build a theory that links subsidiary embeddedness
and subsidiary distinctiveness, explaining how the process of developing its distinctiveness
depends on the nature of its embeddedness. We fulfill this purpose through a single
longitudinal case study of an automobil components manufacture subsidiary.
IESE Business School-University of Navarra
A Preview of the Model
We argue that the subsidiary embeddedness in the MNC is the basis on which its strategy can
be developed, and that the level and type of embeddedness of the subsidiary is a constraint
on its ability to develop strategy as it limits its initiative potential. Our focus in this paper is
not on the local embeddedness of the subsidiary but on its mirror image, the embeddedness of
the subsidiary in the MNC. As our case data shows, the level of subsidiary embeddedness in the
MNC accounts for the capacity of the subsidiary to have its own strategy. We develop this
argument into a theoretical model presented in Figure 1. Subsidiary embeddedness becomes
then an integral part of strategy, as the subsidiary has to eliminate or manage the constraints in
order to develop its own strategy.
Figure 1
A Model of Subsidiary Strategy as Embeddedness
Levels of Embeddedness
Strategic Embeddedness
Headquarters
mandate
P1a
Subsidiary’s
Autonomous
Behavior
P1b
P2b&c
P3
Subsidiary’s
distinctiveness
Capability Embeddedness
P2a&c
Operational Embeddedness
P4
Subsidiaries are paramount to Multinational Corporations’ (MNCs) performance. Our central
argument is that subsidiary strategy can be traced back to its embeddedness consequences, and
we argue that the subsidiary’s strategy will be possible only if built on its embeddedness. The
study of their contribution to the overall MNC has evolved from a merely passive
implementation of head office mandates, towards a more proactive and initiative-taking
strategy (Prahalad and Doz, 1981; Hedlund, 1986; Bartlett and Ghoshal, 1989; Birkinshaw,
1997; Birkinshaw and Hood, 1998; Taggart, 1999; Andersson et al., 2002; Birkinshaw, Hood
and Young, 2005). Underlying this pattern of research is the prevailing idea that subsidiaries
may develop their own strategy by means of two alternatives. On the one hand, the subsidiary
may simply fulfill the responsibilities that come from a specific role assigned by headquarters
(HQ). Subsidiary strategy may be allocated a subsidiary role (Jarillo and Martínez, 1990). On the
other hand, subsidiaries may develop as sources of an MNC’s competitive advantage and
providers of strategic initiatives, following a more autonomous behavioral pattern (Burgelman,
1983). Subsidiary strategy may then be associated with initiative-taking (Birkinshaw and Hood,
1998; Taggart, 1999; Delany, 2000). Therefore, subsidiary strategy can be seen as a complex
balance between HQ’s mandate and autonomous behavior. We view subsidiaries as balancing
these requirements in as much as they manage embeddedness. In line with Nohria and Ghoshal
(1997), we view MNCs as a network of embedded differentiated members. Building on
Granovetter (1985) we use embeddedness as the network of relations with other units of the
MNC and the market in general.
2 - IESE Business School-University of Navarra
The model presented in this paper has four interlinked main elements (see Figure 1). The first
element is headquarters’ (HQs) mandate, which is exogenous to the model. It may vary from a
constrained mandate to a more open guideline shaping the subsidiary’s behavior. Second, the
autonomous behavior of the subsidiary, which depends on the restrictions set by the mandate.
The more leeway the mandate leaves, the more the subsidiary will develop autonomous
behavior. Via this behavior the subsidiary can develop embeddedness, which is the capacity of
the subsidiary to develop social action, which in turn helps the development of subsidiary
strategy. Third, the levels of embeddedness the subsidiary develops through its autonomous
behavior. The resulting three levels of embeddedness in ascending order are operational,
capability and strategic according to the level of social action achieved. Finally, subsidiary
distinctiveness is the outcome variable of the level of embeddedness. Achieving distinctiveness
within the MNC contributes to the subsidiary survival and fosters its development. The extent
to which distinctiveness is accepted by the MNC’s HQ will feed back into the subsidiary’s
autonomous behavior.
The development of a subsidiary named Brakes Spain, using a case study and network analysis,
provides support to develop this inductive model. Due to the delicate nature of the data the
names have been disguised. From the case data, we suggest that the subsidiary can manipulate
embeddedness as part of its strategy, as it can manipulate other resources. Embeddedness is not
only determined by the external factors surrounding the subsidiary, but can also be seen as the
result of the autonomous efforts of the subsidiary to stand out for adding value to the overall
MNC. The three dimensions of embeddedness proposed here, which the subsidiary can effect in
order to increase its strategic importance, are explained next:
Operational embeddedness is viewed here as the set of relations between the subsidiary and
other units of the MNC that concern day-to-day activities. Coordination of worldwide
manufacturing or product launch gives rise to a network of relations across the MNC whose
aim is to keep operational day-to-day activities afloat.
Capability embeddedness is viewed here as the set of relations that comprise the development of
strategic capabilities (Teece, Pisano and Shuen, 1997). Benchmarking, best practices or
knowledge management efforts highlight the competitive proficiency of certain units. The
diffusion of these practices generates a network of relationships among operating units. The
more the subsidiary develops these strategic capabilities, the higher the chances of being called
to contribute. Overall knowledge management processes will integrate the best actors within the
subsidiary to contribute to the different strategic capabilities of the multinational.
Strategic embeddedness is viewed here as the set of relations between the subsidiary and other
units of the MNC that relate to the subsidiary’s participation in the overall strategy of the MNC.
Active participation of subsidiary members in the strategy process may influence the overall
MNC’s strategy through a group of relations generated from the subsidiary. The more
participation there is in the strategic process of the MNC, the more dependencies will be
encouraged among the different units in the MNC, and the higher the level of strategic
embeddedness.
The proposed model posits an active and important role of subsidiaries in breeding relationships
with other members of the MNC. The content of these relationships may range from an
operational nature, through capability content all the way to a more strategic character.
However, it can also go into a decline in the opposite direction losing either strategic or
capability embeddedness due to subsidiary inaction or change in HQ’s mandated constraints.
IESE Business School-University of Navarra - 3
The remainder of the paper is organized as follows: the next section will introduce the research
settings and describe the methods and sources of data. Then we take the reader through the
story of the company introducing both qualitative and quantitative data, by virtue of which we
develop the theoretical model. Finally, in the last section, we discuss the limitations of our
approach, including the relevant boundary conditions, and highlight implications of the model
for future MNC research.
METHODS
This study can better be understood as inductive theory building. We have used a single case
study design to study evolution of subsidiary strategy over a ten-year period. The four sources
of data, including one quantitative, are presented in Table 1. Informal interactions coupled with
participative observation provided the bulk data to construct the story. Participant observation
accounts were the result of one author’s collaboration with the company studied. This type of
data offered depth towards understanding the genesis of strategic behavior in Brakes Spain. In
2003, a second author of this paper collected his qualitative data through interviews in order to
be able to cross check the first author's interpretation. We also collected network data and
carried out semi-structured interviews in 1998 and 2003. A detailed account of this process
together with the research setting follows.
Table 1
Sources of Data
Interviews a
Year
2003
Sub-Total Top
31
Man.
23
Measurement of Social Network c
Middle
Man.
Informants
8
20
1998
37
22
15
18
Total
68
45
23
23 b
Sample
(Population)
Resp. rate
84
(1685)
95%
50
(1500)
96%
Ego-alter
d
pairs
Intern
NW
Extern NW
Corp NW
1827
1.315
232
280
1411
734
322
355
Participant
Observation e
Archival Documents
Level
Examples
Examples
Top Management
- Planning meeting
- Budgeting meetings
- Brainstorming
sessions
- Assessment and
reporting meetings
-
Middle Management - Objective setting
meetings
- Budgeting meetings
- Content developing meetings
- Product
development
meetings
Strategic objectives
Organizational change
plan
Industry studies
Historic company
records
Summaries of strategy
meetings
Communication records
with HQ
Strategic Plans
Total 35
a
Data collection included 68 in-depth one to two hour open-ended, in-person interviews with managers of levels 1,2,3 and 4 from the studied organization. These included two interviews each
time with each of the three subsidiary CEOs. In 2003, two of these CEOs had left the company, but were still interviewed. Other interviewees were: Plant Managers 1 and 2, Head of Purchasing,
Product Testing Manager, Chief Engineering Officer, Human Resources Manager, Client 1 Account Manager, Application Engineering Liaison, Controller Brakes, Spain Quality Manager Purchasing
Manager, Product Testing Manager.
b
5 interviewees, not present in 1998, were interviewed in 2003.
c
The Social Network was measured in 1998 and 2003; both times via a questionnaire that measured work-based relationships as long term patterns of interaction (Freeman, Romney and Freeman,
1987). Questions specifically asked separately for relations with the subsidiary (internal), with the MNC (corporate) and with suppliers or clients (external) (Knoke and Kuklinski, 1982). The sample
was selected using a “reputational approach” (Scott, 1991), where information about the actors with higher level of reputation in the set is considered enough to determine the networks of
relations of a set. Sampling was required due to both the natural inability of respondents to complement a massive questionnaire and the computational limitations of any network software.
d
Evident mismatches and incomplete entries were eliminated resulting in 99.2% of usable data for 1998 and 2003.
e
Participant observation was carried out from 1993 to 2003. Through this ten-year period the frequency of interaction ranged from weekly to monthly. These observations were characterized by
informal interactions, visits to the company and participation in meetings.
4 - IESE Business School-University of Navarra
Research Setting
Brakes Spain, a subsidiary of British, and managed via its Brakes Division, is the critical case
used to inductively develop theory in the present paper. Brakes Spain has undergone an
organic process built around the growth and decline of valuable and distinctive resources as
illustrated in Birkinshaw and Hood (1998). For this reason, we have chosen it as a revelatory
case. Using existing theorization in subsidiary evolution we have inductively developed the
theoretical model set forth in the present paper.
British is a diversified MNC, active in more than 30 countries across Europe, America and Asia.
Brakes Spain belongs to the brake division within the automotive business, Brakes, a world
leader with more than 25 percent volume world market share. Brakes Division is British’s
largest business and most of the leading vehicle manufacturers utilize their components,
operating in 20 countries. Active R&D is a key element for Brakes Division due to the limited
number of customers. Concentration and consolidation of the vehicle manufacturers (Womack,
Jones and Roos, 1991) has limited the number of customers for British worldwide.
Turnover of Brakes Spain was in the region of €350 million in 2001, operating two plants with
1,800 employees. Amongst all Brakes subsidiaries, the Spanish one stands out for three reasons.
First, Brakes Spain has been consistently the best performing subsidiary in the Brakes Division,
showing the highest return on sales, as well as high flexibility and productivity. Second, all
customers’ decision centers are located outside Spain, leaving Spain in a disadvantageous
position compared to other subsidiaries. This forces Brakes Spain to be especially active in
developing its position within the global market and across the MNC. Third and foremost,
Brakes Spain has been not only a net cash contributor but also a source of strategic initiatives
and capability development for the Brakes Division. As such, Brakes Spain has grown to be a
unit of high strategic importance to the MNC.
Data Collection
Participant Observation: Our involvement with the studied company began in 1993 when one
of the authors was invited to take part in a meeting at Brakes Spain. The objective was to
review the company’s situation and develop a detailed plan for implementing the desired
changes. From then until 2003, regular contact with Brakes Spain was maintained at several
levels. This included informal interaction, meetings and discussions with managers throughout
the company. Participant observation aimed at facilitating strategy formation within Brakes
Spain. Regular written notes provided a key resource as a mean of narrating the story. Informal
interactions and meetings varied in frequency from weekly to monthly. Weekly interactions
took place with contacts across hierarchical levels, from factory foremen through managers to
the CEO. The main monthly interactions took place with Brakes Spain’s top management. For
certain periods of time, there was also frequent interaction with four divisional level managers.
Interacting regularly with managers and operational personnel provided data which are not
commonly available, and became the basis for narrating the story and articulating the
theoretical links that explain it.
Interviews: In parallel, and to gain further and formal understanding of the strategic evolution
of the subsidiary, 31 in-depth interviews were carried out in 2003 and 37 in 1998. These
interviews took place from March to December 2003 and March to August 1998. On both
IESE Business School-University of Navarra - 5
occasions, interviews lasted between 45 minutes and 2 hour, and detailed notes were taken. On
both occasions, a protocol was followed as guidance for the sequence of steps to follow and
themes to cover aiming at reliability (Yin, 1994). The interviewees were key officers who had a
role in the evolution of the subsidiary strategy, as shown in Table 1. We interviewed 23
managers directly involved with the formulation and implementation of Brakes Spain’s
strategy. Two researchers were involved in developing the interviews to avoid biased
interpretation and to assure coverage of all relevant issues. Throughout the data collection and
analysis process, we aimed at internal validity, looking for alternative explanations between the
two researchers (Yin, 1994). After each round of interviews, the authors had critical discussions
to validate interpretation (Silverman, 2001). Whenever disagreement arose it was sorted out by
making a phone call to specific interviewees. This entire process aimed at maintaining
reliability and internal validity during qualitative data collection.
Archival data: This included the following documents: first, strategic and organizational
documents, at both national and divisional level; second, organizational charts and
organizational change plans; third, industry studies and competition reports that analyzed the
position of Brakes Spain; fourth, advice and communication records with external consultants;
fifth, presentations and communications records with HQ and other business units, as well as
historic records of the relationship with customers. These documents were key in cross-checking
with the other sources of data.
Social Network Analysis: Our aim was to determine the work-based pattern of relationships
that start at the units that make up Brakes Spain, both internally and externally. We included
every relevant player within the firm in the sampling . The sampling method was the
“reputational approach” (Scott, 1991), where obtaining information about the actors with
the higher level of reputation in a particular set of actors is enough to determine the network of
relations of that set. This criterion proved useful at Brakes Spain, where managers are the
actors with the best knowledge of which agents enjoy the highest level of reputation in
the company. Yet, to avoid “hierarchy bias” (tendency to select one’s subordinates as enjoying
high reputation), the researchers turned to the authors’ knowledge of the firm and its
employees, and selected those with higher levels of reputation among their peers. The final list
was compared to a different one compiled by Brakes Spain’s Human Resources Manager,
adding or deleting members based on a discussion about his/her actual level of reputation. As a
result of this process, 50 managers were selected in 1998 and 84 in 2003. Questionnaires were
handed out in the first wave and administered online in the second wave. Questions were
formatted in a free-recall and fixed-choice design to allow respondents to name any actor
belonging to any unit in Brakes Spain or the MNC with whom he or she had interacted at work
in the last six months. Response rates were 95% and 96%, for 1998 and 2003, respectively. In
both instances, personal calls were made to individuals who were reluctant to respond at first.
In both cases a presentation of aggregate network results was made to different units within the
company, which also helped to clarify our interpretation of the results.
Data Analysis and Reduction
Combining participant observation, interview, archival and network data allowed us to
understand, not only the relationships between entities, but also the content of those
relationships. The analyses of qualitative and quantitative data were carried out separately and
hence they are presented next. Both types of data were linked after the analysis to develop our
theoretical model.
6 - IESE Business School-University of Navarra
Qualitative data: We used various sources of data to develop the Brakes Spain case. For
consistency, we triangulated the information obtained through the interviews with the
longitudinal database derived from our involvement with the company and the access to
archival data (Yin, 1994). To make sense of such a vast amount of data, we focused on
constructing the story making use of narrative to consolidate our understanding. Based on the
record of participants observation, the two sets of interviews and archival data, the story was
written, including the logical links between events. Use of narrative is useful to reduce large
amounts of data into a coherent unit (Miles and Huberman, 1994; Pentland, 1999; Silverman,
2001). Interviews and records of participant observation were organized into a sequential
timeline matrix (Miles and Huberman, 1994). We classified each segment into one of three time
categories, namely, 1993-1998, 1998-2002 and 2002-2003. Next, we devised a written
document that logically collated each segment within each time category. Successive iteration,
between the matrix and the text, and cross-checking with archival data produced a 34 page
draft reporting the story of Brakes Spain. Aiming at reliability, we pursued authentication of
the written case by two independent consultants from Brakes Spain (Yin, 1994). They read,
revised, and validated the narrative. After a few minor modifications the case in narrative form
was deemed an accurate description of the events occurring in Brakes Spain from 1993-2003.
A version of this narrative is presented in the present paper. The analysis that logically links
operational, capability and strategic embeddedness from the story data can be seen in Table 2.
IESE Business School-University of Navarra - 7
8 - IESE Business School-University of Navarra
2002-2003
III. Losing
Ground
1998-2002
II. Gaining
Fitness
“…our production aimed at reducing labor by
mechanization, which reduced time. The way we did it
was so novel that they came from the U.S. to ask for
advice. We had to tell them what they had to change.”
(Int. 5)
Some efficiencies
can be exported
to the rest of the
MNC
“How to manage a plant within an MNC
that requires 8% to 10% it improvement,
well, you have to spend wisely.” (Int. 4)
Examples (Sources of data)
Losing control
over key activities
weakens
capability
“Currently, our advantage rests upon
manufacturing, trying to keep that process
on hand with the entire process that
precedes launching of new products.” (Int.
55)
Constraints
“Not only have they taken away control
from HQ affect over processes and taken all our
executives but they have also made us
execution
reduce personnel by 10 percent, treating
us like any other unit.” (Int. 53)
Distance from key “It is harder for us now, given that we are no longer
activities weakens responsible for key activities, nor do we have control
over them and global responsibilities are far away
capabilities
from us.” (Int. 41)
“At the same time we had developed a flexible
structure to launch new products. Sales and
engineering were tightly knitted between themselves
and with the customers.” (Int. 22)
“Our control over marketing, sales and engineering...it
was the integration of all those activities that made us
strong at launching new products and gave us an
advantageous position across the MNC.” (Int. 41)
Gaining control
over key activities
enhances their
position
“Our process of launching new products
bases its flexibility on the direct control of
all the related functions, from quotation to
mass production.” (Int. 18)
Coordination
efficiencies
Manufacture
efficiently
“We managed to develop a renowned engineering
department, that would be present in any group
product development all across the multinational.” (Int.
21)
Creating a core
capability
enhances power
with the MNC
“No other unit wanted to manufacture
component X. It was too complex. We
accepted the challenge at the end of 1997.
As a result, we now supply that
component to the whole group and with a
decent margin.” (Int. 10)
Challenges
and risks taken
prove
rewarding
“We wanted to be someone within the group and be
respected, not because of cheaper labor costs but
because we did well and are important to the group.”
(Int. 1)
Operational
excellence
Description
High aspirations
for the future
Examples (Sources of data)
...until 1998, our role was to execute
orders from HQ and other units, we were
simply manufacturers, with simple plants
and a marginal role within the group.”
(Int. 1)
Description
Mainly
efficiencies in
production
standards and
levels
Stage
Capability Embeddedness
I. Obeying
Orders
1993-1998
Operational Embeddedness
Levels of Embeddedness
Table 2
Description
Examples (Sources of data)
“From the subsidiary viewpoint, one moves in a
political environment, within which you have to
get what you want without upsetting others. So
far nobody has felt that way with our moves,
since Brakes Spain was not important within the
group.” (Int. 46)
“Our structure was even copied in the U.S. From
HQ the mandate was to follow the structure we
had invented. We were recognized as skilful
people then.” (Int. 31)
“…if we wanted to avoid being relocated to a low
cost country, it was necessary to look for valueadding alternatives. We decided that Brakes
Spain would contribute significantly to the whole
group.” (Int. 1)
Losing key actors
diminishes
strategic potential
“When they created the global account manager,
the role of sales manager was devalued. We had
no control over the customer, not even at product
development level. Obviously, our sales manager
decided to leave.” (Int. 38)
Lost interaction
“When the division decided to centralize key
engineering activities and its relationships with
accounts for
vanished strategic clients, we lost that control and, consequently,
our flexibility for launching new products.” (Int.
pursuit
31)
Interaction with
“Our structure is flat, we base our contribution on
key actors creates the relationship amongst our mangers, with HQ,
potential strategy
with the customers and with our plants.” (Int. 26)
Best practices
adopted
elsewhere provide
reputation
Establishing
relationships and
managing them
Contributing to the
MNC is seen as
the way forward
Strategic Embeddedness
Network Analysis: The issue which recurred most during our interaction with managers was
the idea that Brakes Spain was losing contact with different counterparts of the business (other
units of the MNC, units of the subsidiary itself, and clients) during the period of study. In order
to validate this claim, we performed a network analysis of the relationships starting at Brakes
Spain and oriented towards an ‘outside’ sphere, namely, customers, other operating units within
the division and the divisional headquarters. The network was measured in 1998 and in 2003.
Structural variables (Wasserman and Faust, 1999) were measured as work-based interactions.
Here, we studied actors from two different sets tied by a work-based relationship, one
consisting of a sample of employees at Brakes Spain, and a second larger set consisting of
employees of the same company plus employees of other units of the MNC, as cited by the
interviewees. Actors in the first set are “senders” of the relation (egos), while those in the other
set are “receivers” (alters). The second set is opened to who ever senders mention. The resulting
network is one made up by ties surrounding the sampled individual units, which can be
separated into three more concrete networks. First, the Internal Network made by the
relationships between and across units of Brakes Spain. Second, the Corporate Network made
by the relations of Brakes Spain with alters, who are members of the MNC. Third, the External
Network formed with customers and suppliers. An illustration of these appears in Figure 2. The
Internal Network shows the relationships among units within the subsidiary, plant departments
or local headquarters departments. The Corporate Network is a two-mode network showing the
relationships of the departments within the subsidiary with different departments in other MNC
units, including Brakes’ division headquarters. The resulting network analysis produced a final
database of 1,411 pairs of ego and alter tied by work-based relation entries in 1998 and 1,827
entries in 2003. This was divided into Internal, Corporate and External Networks as can be seen
in Table 3. We performed the analysis at the aggregate functional level since we were interested
in the relationships between functions rather than between individuals. Next, data were
analyzed on cross-tabs, each cell representing the frequency of work-based interactions
between ego and alter. To identify the differences between the two networks we calculated the
proportion of relations in each network to subsequently test if their differences were significant.
The results of such calculations can be seen in Table 3. In addition, we block modeled the two
mode Networks (Borgatti and Everett, 1997) and calculated centrality and density measures for
the External Network as can be seen in Table 4. All calculations were done using UCINET
software for Social Network analysis.
IESE Business School-University of Navarra - 9
Table 3
Evolution of Relations in Brakes Spain
1998 a
Number of Relations
All Ego-Alter Relations (I,II & III)
1,183
I. Internal Network: Operational
and Strategic Relations
Number of Relations
p̂2
pˆ 1 ≠ pˆ 2
1705
1
734
0.62
1315
0,77
(+) ****
1) HQ Spain-Plants
106
0.09
185
0.11
(+)
2) Plants-HQ Spain
78
0.07
103
0.06
(-)
3) HQ Spain-HQ Spain
322
0.27
408
0.24
(-)
4) Plants-Plants
228
0.19
619
0.36
(+) ****
325
0.27
292
0.17
(-) ****
1) HQ Spain-MNC Plants
156
0.13
136
0.08
(-) ****
2) HQ Spain-MNC HQ
118
0.10
89
0.05
(-) ****
3) Plants Spain-MNC HQ
39
0.03
14
0.01
(-) ****
4) Plants Spain-MNC Plants
12
0.01
53
0.03
(+) ***
124
0.10
98
0.06
(-) ****
114
0.10
58
0.03
(-) ****
12
0.01
40
0.03
(+) **
III. External Network: Operational
and Strategic Relations
1) HQ Spain-Clients
2) Plants Spain-Clients
b
p̂1
1998 vs. 2003b
1
II. Corporate Network: Operational
and Strategic Relations
a
2003 a
p̂1 and p̂2
are the proportions of the total relationships in the category.
Significance of difference of proportions has been calculated using a normal distribution.
**** p<0,0001; *** p<0,001, ** p<0,002.
10 - IESE Business School-University of Navarra
Table 4
Corporate Nertwork Brakes Spain towards Brakers:
Density Image Matrixes 1998 and 2003
1998
0.196
Two- mode network density
2003
0.156
Block Modeling 1998
Brakes Division
A
B
X
1
1
Brakes
Spain
Y
0
X Group (Brakes Spain actors): HQ-Product Engineering, HQ-Marketing and Sales, HQResearch and Development, HQ-Purchasing. Plant Process Engineering
Y Group (Brakes Spain actors): HQ-Testing, HQ-Controlling, HQ-Purchasing, Plant
Quality, Plant Purchasing, Plant Controlling, Plant General Management, Plant
Manufacturing, Plant Human Resources, HQ-Quality.
0
A Group (Brakes actors): HQ-Product Engineering, HQ-Research and Development, HQManufacturing, HQ-Purchasing, HQ-Product Engineering. Plant Product Engineering,
Plant Purchasing.
B Group MNC (Brakes actors): HQ-Marketing and Sales, HQ-Testing, HQ-Human
Resources, HQ-Control, HQ-General Management, HQ-Quality, HQ-Product Engineering.
Plant Quality, Plant Controlling, Plant General Management, Plant Testing, Plant
Research and Development, Plant Process Engineering, Plant Logistics.
0
X Group (Brakes Spain actors):HQ- Product Engineering, HQ-Marketing and Sales, HQResearch and Development, HQ-Testing, HQ-Quality, Plant Quality, Plant Process
Engineering.
Y Group (Brakes Spain actors): HQ-Controlling, HQ-Human Resources, HQ-General
Management,
HQ-, Purchasing, Plant General Management, Plant Manufacturing, Plant Human
Resources, Plant Controlling, Plant Logistics.
0
A Group (Brakes actors): Plant Quality, Plant Marketing and Sales, Plant Product
engineering, Plant Process Engineering.
B Group (Brakes actors): HQ-Quality, HQ-Marketing and Sales, HQ-Purchasing, HQControlling, HQ-General Management, HQ-Testing, HQ-Human resources, HQ-Research
and Development,
HQ Engineering, Plant Controlling, Plant Testing, Plant manufacturing, Plant general
management, Plant Human Resources, Plant Research and Development, Plant
Logistics.
Block Modeling 2003
Brakes
A
Division
B
X
1
Brakes
Spain
Y
0
For the Brakes divisional level actors, plant departments have been consolidated. For instance, Plant Quality is the aggregation
of all quality departments of Brakes plants around the world consolidated as one single actor.
Brakes Spain Towards Brakes: Density Matrices per Block
2003
1998
X
A
0.722
B
0.234
Y
0.169
0.098
A
B
X
0.515
0.149
Y
0.070
0.096
Number of alter per year
1998
XA
2003
XB
Total
XA
HQ
Plant
HQ
Plant
HQ
Plant
HQ
Plant
HQ
5
2
7
7
12
9
0
4
9
XB
Plant
7
HQ
9
Total
Plant
11
To be read: In 1998, the number of departments from Brakes division headquarters to which actors in
Block X from Brakes Spain were related, was 5.
IESE Business School-University of Navarra - 11
Figure 2
Network Evolution
1998
2003
Internal Network
Corporate Network: Two-mode network representation where MNC’s functions are one dimension
and subsidiary functions are the second dimension. Squares represent MNC’s functions and circles
represent the subsidiary’s functions
BRAKES SPAIN: THE HISTORY OF ITS STRATEGY
The story of Brakes Spain is one of a regular subsidiary that was able to gain significance in
the strategic development of its mother company. However, Brakes Spain's strategic importance
was jeopardized as the MNC centralized key activities. From 1993 to 2003 Brakes Spain
suffered initially a rise and then a decline in its strategic importance. Firstly, Brakes Spain
moved from being a mere plant to becoming a strategic player within the global company.
Then, it started to lose strategic relevance in such a way that its managers perceived the
subsidiary was going back to its mere operating role within the MNC. We summarized this story
into three different stages or time categories that we have called: Obeying Orders, Gaining
Fitness and Losing Ground.
I: Obeying Orders
Prior to 1998, the business strategy of the MNC British had been one of optimizing operations,
reducing costs and minimizing delivery time in a globalized marketplace. The globalization
12 - IESE Business School-University of Navarra
process carried out by the MNC had led Brakes Spain to a situation where its customers’
decision centers were dispersed worldwide far from its national operations. Moreover, by 1993,
Brakes Spain had few local customers and neither proprietary technology nor a team to
develop it. This situation stood as a clear disadvantage vis-à-vis other European subsidiaries,
which either had relationships with local customers or enough R&D to cope with developing
new products. Managers of Brakes Spain felt its manufacturing operation was a candidate for
relocation to a lower labor cost country. In order to overcome this dismal future, the subsidiary
decided to design a strategy that allowed Brakes Spain to gain distinctiveness within the
division. The subsidiary’s CEO summarized this notion as:
“If we were to avoid the plant’s relocation, something that was happening to other plants
of the group, it was vital for Brakes Spain to be able to add value to the MNC; we
decided to transform our unit in order to make it able to make a greater contribution to
the division; we wanted to base our prestige in efficiency, the best manufacturing and
high flexibility.”
The strategy designed to overcome the ‘Obey Orders Stage’ was based upon three pillars. First,
the development of a strong local HQ in order to improve coordination and knowledge
transmission among plants. Fulfilling this function, Brakes Spain expected to serve as a filter
between its plants and other units of the MNC. Second, they invested in R&D subsidiary
capabilities, giving emphasis to the engineering department. Several young engineers were
recruited with the objective of generating novel manufacturing processes, more efficient plant
layout designs, and better product applications. This investment was also expected to indirectly
give Brakes Spain a better position compared to similar units in countries such as France and
Germany. Should this plan work, Brakes Spain would be able to establish immediate and fluid
relationships with both the engineering departments of other units of the MNC and those of its
clients. Finally, the strategy aimed at developing the sales department, in order to bring
customers closer to Brakes Spain.
In order to compensate for their dull geographic position, Brakes Spain decided to push its sales
department to develop the Asian market, hiring international and well-prepared managers who
could be in direct touch with customers and engineers of other units of the MNC. They decided
to push the production department and their plant managers in order to bargain internally
within the MNC to acquire projects and products.
The overall strategy was brought up by the subsidiary, in the form of a suggestion, to the
divisional head. The argument used was that the integration of the subsidiary within the group
activities, being closer to the customer and the division, would allow Brakes Spain to produce
faster, respond quicker and add more value to the division as a whole. Nevertheless, the process
followed to gain the division’s authorization was quite eclectic. In words of the 1993-1998 CEO:
“…Sometimes you negotiate these things with head office, some they grant and some
they don’t… sometimes you just don’t make a proposal, you simply act and head office
endorses it after the fact”
The results of the implemented strategy were evident in 1998, when, as a consequence of sound
management by the CEO of Brakes Spain he was promoted to a higher position within the
MNC. Additional results attributed to this strategy determined the second stage of Brakes
Spain’s strategy.
IESE Business School-University of Navarra - 13
II. Gaining Fitness
By 1998, Brakes Spain had turned the previous situation around. With a share of 11% of total
MNC’s turnover, it was the largest subsidiary of the Brakes Division in terms of cash flow
contribution to the division. During the 1998-2002 period, several managers from the Spanish
subsidiary were promoted to divisional functions and responsibilities. This reveals the
contribution potential of the unit within the MNC. Managers’ promotions were driven by the
perception at Brakes Division that Brakes Spain had become more capable of contributing
strategically to the global organization, due to its managers’ expertise in operations and
business processes. For example, Brakes Spain was able to put a prototype into mass
production in record time. What was more, from the selling perspective, customers from distant
countries such as the U.S. or Japan would specifically asked to be served by Brakes Spain.
Similarly, after a global benchmark process, the MNC decided to clone the task forces model
designed by Brakes Spain across the rest of the subsidiaries to launch new products.
Furthermore, the MNC decided to ‘translate’ the layout designs developed by Brakes Spain to
the U.S. Additionally, after negotiating with the Brakes Division, plant managers at Brakes
Spain were able to gain for the Spanish subsidiary the production of certain parts that were
turned down by other units of the MNC. This negotiation left Brakes Spain specialized and
responsible of the production of high-margin brakes-system parts within the group.
In sum, Brakes Spain achieved a high reputation across the MNC based upon its manufacturing
excellence, its best selling team, its flexibility at launching new products, its well-known
engineering team and the fact of being the largest subsidiary of Brakes in terms of net cash
contribution. Consider, for instance, the effect that Brakes Spain had on HQ via its involvement
in the Asian expansion strategy. In different European meetings, HQ declared the importance of
serving Asian customers. However, none of the MNC subsidiaries felt confident enough to
overcome problems derived from cultural and language differences and geographic distance. At
a certain point in time, the strategy of expanding to Asia seemed impossible. Nevertheless,
Brakes Spain pushed its commercial department harder and was able to start projects with one
client in Japan and one in Korea, which was the first step to making the Asian expansion
feasible.
The strategy proposed in 1999 at Brakes Spain aimed to keep and strengthen the advantageous
position achieved. It focused on keeping the state-of-the-art organization and mass production
techniques, but added a search for excellence in engineering, to strengthen the identity of the
plants, foster good relationships with the workers as well as develop international executives
who could aspire to positions at Brakes Division. More than anything else the strategy
strengthened their R&D department by looking for recognition within the group while helping
develop the Asian expansion. Nevertheless, the future was going to prove different to that
intended.
III. Losing Ground
The key change occurred by mid-1998, when a new divisional head was appointed, with
subsequent repercussions. His main self-imposed strategic goal was to attack the Asian market
through the European operations given their level of technology, efficiency, reliability and
geographical location (closer than the U.S., for example). Meanwhile, from the divisional
viewpoint of Brakes, it was thought that, in order to achieve this goal, a higher level of
centralization was needed. Consequently, by 2002 purchasing, new product launches,
14 - IESE Business School-University of Navarra
engineering applications processes and sales activities were centralized. Additionally, the
division also imposed a certain level of rationalization at the manufacturing level, trying to
leverage specialization of plants. The argued rationale behind these decisions was to keep
control over key business activities, to better coordinate R&D investment among units and to
obtain lower manufacturing costs.
This set of measures scaled back the relationships that Brakes Spain had established with
customers, as well as with other units of the multinational. The consequences damaged Brakes
Spain’s autonomous strategic behavior. As global managers and functions were established,
local managers and functions started to vanish. Key decisions started to be taken by global
managers and only low-level decisions were kept in-country. For instance, under the new
structure, Brakes Spain was not allowed to contact a customer directly when the customer
decided to include a small change in the design of a new product or application. That function
was now the responsibility of the new Global Account Manager, who, in turn, had to
coordinate with global project leaders. Apart from losing contact with customers and managers
from other units of the MNC, Brakes Spain’s field was forced to narrow its strategy towards the
achievement of operational excellence. The strong pressure to reduce headcount was unheard of
in an organization that had always been focused on customer strategy and growth. In the
aftermath, the strategic role played by Brakes Spain vanished under centralized activities such
as marketing, sales, new product launches and R&D. In the words of the 2002-2003 CEO:
“A great part of our reputation within the group was due to our closeness to the customer
and the flexibility of our product launching teams. From 2000, when European
representatives were appointed to these activities, it is they who were to address the
customer, while we could barely reach the customer directly…now we do not control the
business and we have gone back to pre-1998: we are simply manufacturing plants.”
A MODEL OF SUBSIDIARY STRATEGY AS EMBEDDEDNESS
The story of Brakes Spain shows the transition of a marginal unit that shaped its destiny. It did
so by taking an active role in building valuable relationships. The subsidiary managers had
decided to design a strategy that allowed the unit to gain distinctiveness within the MNC.
However, a constraining level of centralization promoted at divisional level in mid-2000
reduced the strategic space that Brakes Spain had enjoyed. It lost its level of autonomy through
the centralization of resources and decisions, which counterbalanced the subsidiary’s attempts
to govern its future.
Underlying this process is a change in the network of relationships from Brakes Spain towards
other units of the MNC and its clients. First, it responded to operational integration by
developing a network of relations, generating Operational Embeddedness. Second, it leveraged
its operational efficiencies into distinctive capabilities establishing relations across the MNC,
generating Capability Embeddedness. These capabilities, in turn, helped to develop a strategic
role, as Brakes Spain championed new markets and developed new products and processes
copied across the MNC, reaching Strategic Embeddedness. In the last stage, they lost control
over key business processes through centralized key activities and went back to Operational
Embeddedness. Each embeddedness type was built on the previous one. Next, we will describe
each of the constructs that make the theoretical model proposed that emerged from the study
data as shown in Figure 1.
IESE Business School-University of Navarra - 15
Operational Embeddedness
By operational embeddedness we mean the set of relations between Brakes Spain and the rest of
the MNC that involve operational and manufacturing activities. As can be seen in Table 2, for
Stage I, the subsidiary concentrated in developing efficiencies in operation, manufacturing and
coordination amongst plants. In a similar fashion towards the end of this study, operational
embeddedness was prevailing. This is reflected in the significance of the relationships among
Spanish plants (36% of all the relations) and the increased significance of the relations between
the Spanish plants and those of the customers (3% of all relations) (See Table 3). Operational
embeddedness is highlighted in table 4 as most of the relations appearing in the only non-zero
block in 2003 are towards units in other divisional plants (block density = 0.515). As the data in
this study suggests, operational embeddedness implies more relationships among plants within
the division and a lower level of relations between local HQ and divisional HQ.
Capability Embeddedness
Capability embeddedness is the set of relationships resulting from efforts to further develop
MNC’s strategic capabilities. Brakes Spain was first able to develop operational embeddedness
through its increasing participation both in best-practice transfer from Brakes Spain, and its
participation in general best-practice workgroups within the MNC. Best-Practice transfers were
informally developed by Brakes Spain by sharing their know-how. Work group relations
emerged as other units in the MNC asked for them. Next, the subsidiary was able to develop its
own capability in product development through recruitment of R&D and engineering staff.
Coupled with development, they built a sales department that would work in tandem with R&D
and Engineering, catering to specific customer requirements. This structure, together with
manufacturing excellence, increased its position to capability embeddedness as shown by the
data in Table 2. This observation is supported by the network measurement in 1998 of a high
number of interactions between Brakes Spain and customers, as shown in Table 3.
The non-zero value of the XB block in the 1998 block model highlights the existence of
relationships between Brakes Spain and different departments in Brakes Division, both at the
plant and the HQ level. The fact that all those relations came from the Brakes Spain head office
illustrates how those were purposefully managed by top management in Brakes Spain’s head
office (see table 4). Through this process Brakes Spain became an integral part of strategic
capability development in the division.
Strategic Embeddedness
By strategic embeddedness we mean the set of relations that Brakes Spain established with
other units of the MNC and with Brakes Division. The fact that Brakes Spain had consistently
good performance within Brakes, had developed distinctive capabilities and gained customer
recognition, made them a significant player within the MNC. They were perceived as a set of
skillful executives who had transformed their organization (see Table 2). Their intention of
developing a distinctive character to avoid being relocated had reached fruition when their
structure to serve customers was copied across the MNC, but more importantly when they acted
as the flagship to break into the Asian market. This pattern is also shown by the social network
analysis. Table 4 shows that the number of alters in 1998 was higher among Brakes’ HQ
16 - IESE Business School-University of Navarra
departments than among other subsidiaries’ departments. Their relations with the MNC at the
level of HQ and Plants were significantly higher in 1998 as shown in Table 3. Moreover, the
number of relationships of the entire Corporate Network was significantly higher in 1998, when
Brakes Spain reached its strategic embeddedness peak. Overall, strategic embeddedness was
developed through the increased presence and relevance in the MNC strategic forums, which is
shown in different patterns of relations within the MNC.
HQ Mandate versus Subsidiary’s Autonomous Behavior
For the subsidiary to present autonomous behavior, the HQ mandate must allow it. The
evidence from the case presented here clearly shows that HQ allowed Brakes Spain to behave
autonomously (see Table 2). Perhaps more important was that the subsidiary capitalized on this
leeway, building a vast network across the MNC and with external entities. Table 4 shows that
there was a higher density in 1998 between Brakes Spain strategic departments and the MNC
than in 2003, illustrating a retrenchment of subsidiary behavior in its international relations.
Between 1998 and 2003, Brakes Spain intensified the relations among its units while
significantly decreasing the effort to relate to other units in the MNC as the percentage of
overall relations between Brakes Spain and Brakes or its customers decreased significantly (see
Table 3). This insight can be summarized in the following propositions:
Proposition 1a: Restrictive mandates from HQ are negatively associated with the subsidiary’s
autonomous behavior.
Proposition 1b: Autonomous behavior is positively associated with the subsidiary’s generation
of capabilities and strategic levels of embeddedness.
From Operational to Capability to Strategic Embeddedness
As the subsidiary was leeway allowed to develop relationships with customers and other units
of the MNC, Brakes Spain was first able to develop operational embeddedness, engaging in
quality control with other units of the MNC (see Table 2). Next, the subsidiary through
recruitment of R&D and engineering staff was able to develop its own capabilities, increasing
its relations with the MNC capability embeddedness. This sequence suggests that constructing
embeddedness is a process in which establishing one level of embeddedness makes possible to
build the next level. In the case of Brakes Spain, initial operational efficiency and establishing
a strong Internal Network led to building new capabilities and establishing a Corporate
Network. Then, the subsidiary was able to develop Strategic Embeddedness by gaining influence
in the overall decision making within Brakes. As the subsidiary grew into developing its own
strategy, it gained distinctiveness within the MNC. Network data for 1998 shows that the
External Network was more dense and with a significantly higher proportion of relations (see
Tables 3 and 4). As data shows in Table 5, operational embeddedness led to capability
embeddedness and then to strategic embeddedness. The argument of the model presented is that
in order to move to one you have to have proficiency in the previous level of embeddedness.
Conversely, as shown in Table 5, when centralization of activities was imposed, the network of
relations was scaled back for Brakes Spain. It is significant that the number of alters in Brakes
Spain’s network from Brakes’ HQ decreased from 12 to 9, while the relations with alters
belonging to other MNC plants increased from 9 to 11 (see Table 4). Moreover, most of the
IESE Business School-University of Navarra - 17
relationships to HQ’s functions came from their plant equivalents. The decreasing role of the
Brakes Spain head office in this last phase of the study illustrates the way back to operational
efficiency. Operational embeddedness emphasizes “operational excellence”. This term was used
by the 1993-1998 CEO as his motto and was again reclaimed by the 2000-2003 CEO as the
instruction to his local managers. Operational excellence is reflected in the higher importance
of 2003 relations within Brakes Spain (77% of all relations in Table 3). Once again Brakes
Spain was left with operational embeddedness. This observation can be summarized in the
following propositions:
Proposition 2a: Establishing operational embeddedness is a necessary condition for the
subsequent achievement of capability embeddedness.
Proposition 2b: Establishing operational embeddedness is a necessary condition for the
subsequent achievement of strategic embeddedness.
Proposition 2c: A restraining mandate from HQ is negatively associated with the achievement of
capability or strategic embeddedness.
Table 5
Interaction of Roles of HQ and Subsidiary Determine Type of Embeddedness
Period
I. Obeying
Orders
HQ’s
Stance
Indifferent
Brakes
Spain’s
Stance
Operational
effectiveness
Relations
of Brakes
Spain
Subsidiary
Distinctiveness
One of the plants is granted the status of Global
Product Center by HQ. The reasons are: high know
how, high dynamism between subsidiary’s units,
control on manufacturing and training
Internal
Low
Manufactured products are designed in Germany,
the U.K. and France. The local engineering team is
skilled, but only in solving technical problems, not in
design or training.
1993-1998
II. Gaining
Fitness
Open
1998-2002
Open
III. Losing
Ground
2002-2003
a
Restrictive
Capability
development
Internal
and
corporate
Strategic
development
Internal,
corporate
and
external
Left only with
operational
effectiveness
Examples of Significant Events a
Internal
Medium
High
Medium
Increased investment in engineering, basically in
recruiting new talent for new products. This was in
part negotiated with HQ, but mostly Brakes Spain
preferred to say “sorry” rather than to ask for
permission in a highly political environment. Brakes
Spain is relatively small, so it is not perceived as
troublesome at HQ.
Coupling R&D and sales, Brakes Spain is able to
work developing product with the customer. The
intention is to launch new products to satisfy the
customer, maintaining excellence in operations and
in processes. Accepted by HQ, the intention is
realized to an extent such that Brakes Spain
outperforms all other units of the MNC.
Centralization of engineering together with
centralization of relationships with customers at HQ
leaves Brakes Spain defenseless. Joint action of
sales and engineering is copied from Brakes Spain
to HQ, as the best way to serve the customer and
most employees go to HQ.
Key events have been extracted from interview, participant observation and archival data to deduce the chain of evidence.
18 - IESE Business School-University of Navarra
Levels of Embeddedness as the Subsidiary Strategy
All in all, the evidence suggests that Brakes Spain gained distinctiveness within the MNC via
generation of embeddedness and then, lost it by mandate of HQ (see Table 5). High and medium
distinctiveness are associated with capability and strategic embeddedness. Network analysis
strengthens the evidence suggesting that decline in subsidiary distinctiveness was forced
through the loss of relations within the corporate and External Network (see Table 3). Similarly,
the previous strategic development suggests that the mechanism used by Brakes Spain to
develop strategy was an increase in relations. The relationships that the subsidiary developed
with the MNC and customers were the vehicle for developing their own strategy. The next
proposition is the synthesis of this observation.
Proposition 3: Capability and Strategic embeddedness are positively associated with the
subsidiary strategic distinctiveness within the MNC.
The situation for 2003 is radically different to that of 1998, when Brakes asked the Spanish
subsidiary to replicate its new product launch organization for the whole MNC. Evidence in
Table 5 suggests that Brakes Spain lost its distinctiveness as it lost its capability and strategic
embeddedness. While these two levels were present, their strategy reinforced autonomous
behavior encouraging new product development and the search for new business. When HQ
curtailed their strategy and enforced subsidiary compliance instead of subsidiary
distinctiveness, the feedback did not encourage autonomous behavior. This observation is
synthesized in proposition 4.
Proposition 4: The acceptance of subsidiary’s level of distinctiveness by the MNC’s HQ is
positively associated with higher levels of autonomous behavior.
COMMENTS AND CONCLUSIONS
We have developed a model of subsidiary strategy that presents embeddedness as the key driver
for the subsidiary to generate distinctiveness within the MNC. We have also identified three
levels of embeddedness that can create distinctiveness for the subsidiary within the MNC. Our
results complement existing research in subsidiary strategy (Birkinshaw and Hood, 1998;
Taggart, 1999; Andersson et al., 2002), which identifies the ability of the subsidiary to develop
strategy.
For Andersson et al. (2002) external embeddedness is critical for subsidiary strategy
development and does have a significant impact on subsidiary performance. We complement
their findings; we add to them the criticality of internal embeddedness. To their claim regarding
external embeddedness, our paper adds the importance of three different levels of internal
embeddedness to the development of subsidiary distinctiveness.
Birkinshaw and Hood (1998) emphasize the importance of interaction between the development
of subsidiary capabilities and HQ’s mandate in explaining subsidiary evolution. The model
presented in this paper builds on their development by emphasizing that the subsidiary
capability evolution flows through different levels of embeddedness, which respond not only to
HQ’s mandate but also to the subsidiary’s autonomous behavior. Operational embeddedness
grows parallel with the development of basic operational capabilities. The integration of the
multinational’s worldwide operations implies a minimum level of operational capabilities for
IESE Business School-University of Navarra - 19
global coordination to work efficiently. When the subsidiary has achieved a certain level of
integration, the development of strategic capabilities allows to contribute to the worldwide
learning effort emphasized by Bartlett and Ghoshal (1989), increasing their level of capability
embeddedness. As recognition grows around the MNC, the subsidiary might start to gain a
voice in different issues, achieving a certain level of strategic influence through the
development of strategic embeddedness. As such, the evolution of the capabilities of the
subsidiary parallels the evolution through the three different levels of embeddedness identified
in this paper. The model presented here also acknowledges the critical relationship between
headquarters and subsidiary (Taggart, 1999; Andersson et al., 2005), by including it as a
triggering element. Moreover, as the different stages of embeddedness are developed, subsidiary
distinctiveness influences further autonomous behavior.
The influence of the network position of firms on their performance has been widely recognized
(Zukin and DiMaggio, 1990; Uzzi, 1996; Dyer and Singh, 1998; Håkansson and Snehota, 1998;
Gulati et al., 2000; Rowley et al., 2000; Hung, 2002). Birkinshaw and Morrison (1995) translate
this view into the multinational context emphasizing the importance of lateral relationships
between subsidiaries, that have been found to have a significant influence on performance
(Subramaniam and Watson, 2006). This paper applies this view to the MNC network view
(Ghoshal and Bartlett, 1990; Andersson and Forsgren, 1996) identifying network embeddedness
as the foundation on which the subsidiary can build its distinctiveness within the MNC. The
model in this paper emphasizes that these lateral contributions can be designed in order to
become a specialized contributor, initiating non-specified mandates. It also explains how these
interdependencies are created and introduces a logical sequence for embeddedness evolution.
The model presented here centers on the ability of the subsidiary to develop its own strategy,
and in doing so, affecting MNC’s norms and, particularly, its strategy. Although the power to
design a strategy for the subsidiary may be breed in HQ, conflicting logic or interests from the
subsidiary may reshape both the subsidiary and the MNC strategy. Social action by the
subsidiary is not only shaped and conditioned by the context of action, determined by HQ and
the action of other subsidiaries, but also by the dynamic element of social action itself. If the
subsidiary is able to exercise strategic choice by developing an active role, this proactive
behavior may affect not only the subsidiary but may expand its effect to the whole MNC. This
may be particularly important when the strategy promoted by the MNC endangers the existence
of the subsidiary itself. The overall level of distinctiveness a subsidiary may be capable of
generating will depend on how the subsidiary manages the three types of embeddedness.
Caution must be exercised in discussing the results of the present study. First, it has been
carried out within a Spanish context and cultural elements could be partial explanation of our
findings. Second, Brakes’ network effect tends to be industry specific (Rowley et al., 2000), and
a single case only addresses one industry. Third, being a single case study, conceptual
generalization has been attempted through the model presented. Although we have used
quantitative data to supplement our analysis, the use of qualitative data entails interpretation.
While this approach is very powerful for developing a theory, it does not provide unequivocal
results. Although known precautions have been taken to strengthen validity and reliability,
further quantitative designs in future research can strengthen the framework and concepts
presented here.
All in all, while the literature on subsidiary strategy has traditionally value in focused on the
role it plays within the multinational company, no work, to our knowledge, has operational zed
the concept of embeddedness in concrete strategic and organizational issues. We introduce a
new and critical component for subsidiary strategy. The three dimensions of embeddedness that
20 - IESE Business School-University of Navarra
we propose correspond to three basic dimensions that result in subsidiary distinctiveness.
Although we do not claim to reduce all components of subsidiary strategy to managing its
embeddedness level, the concept of developing embeddedness to gain distinctiveness explains a
process through which subsidiaries can develop a strategy.
IESE Business School-University of Navarra - 21
References
Andersson. U. and M. Forsgren (1996), “Subsidiary embeddedness and control in the
multinational corporation," International Business Review, 5, (5), pp. 487-508.
Andersson, U., I. Björkman and M. Forsgren (2002), “Explaining subsidiary network
embeddedness: The impact of headquarters control mechanisms," (Working paper),
Andersson, U., I. Björkman and M. Forsgren (2005), “Managing subsidiary knowledge creation:
The effect of control mechanisms on subsidiary local embeddedness," International
Business Review, 14, (5), pp. 521-538.
Bartlett, C. and S. Ghoshal (1989), “Managing across borders: The transnational solution,”
Harvard Business School Press, Boston, Massachusetts.
Birkinshaw, J. (1997), “Entrepreneurship in multinational corporations: The characteristics of
subsidiary initiative," Strategic Management Journal, 18, (3), pp. 207-229.
Birkinshaw, J. and A. Morrison (1995), “Configurations of strategy and structure in subsidiaries
of multinational corporations," Journal of International Business Studies, 26, (4), pp. 729754.
Birkinshaw, J. and N. Hood (1998), “Multinational subsidiary evolution: Capability and charter
change in foreign-owned subsidiary companies,” Academy of Management Review, 23,
(4), pp. 773-795.
Birkinshaw, J., N. Hood and S. Young (2005), “Subsidiary entrepreneurship, internal and
external competitive forces, and subsidiary performance," International Business Review,
14, (2), pp. 227-248.
Borgatti, S. and M. Everett (1997), “Network analysis of two mode data," Social Networks, 19,
pp. 243-269.
Burgelman, RA. (1983), “A process model of internal corporate venturing in the diversified
major firm," Administrative Science Quarterly, 28, (2), pp. 223-244.
Delany, E. (2000), “Strategic development of the multinational subsidiary through subsidiary
initiative-taking," Long Range Planning, 33, (2), pp. 220-244.
Dyer, J. and H. Singh (1998), “The relational view: Cooperative strategy and sources of
interorganizational competitive strategy," Academy of Management Review, 23, p. 660.
Freeman, L., AK. Romney and SC. Freeman (1987), “Cognitive structure and information
accuracy," American Anthropologist, 89, pp. 310-325.
Ghoshal, S. and C. Bartlett (1990), “The multinational corporation as an interorganizational
network," Academy of Management Review, 15, (4), pp. 603-625.
Granovetter, M. (1985), “Economic action and social structure. The problem of embeddedness,"
American Journal of Sociology, 91, pp. 481-510.
Gulati, R., N. Nohria and A. Zaheer (2000), “Strategic networks," Strategic Management Journal,
21, (3), pp. 203-215.
22 - IESE Business School-University of Navarra
Håkansson, H. and I. Snehota (1998), “No business is an island. The network concept of
business strategy," Scandinavian Journal of Management Studies, 4, (3), pp. 187-200.
Hedlund, G. (1986), “The hypermodern mnc: A heterarchy?," Human Resource Management, 25,
(1), pp. 9-35.
Hung, SC. (2002), “Mobilising networks to achieve strategic difference," Long Range Planning,
35, pp. 591-613.
Jarillo, C. and J. Martínez (1990), “Different roles for subsidiaries: The case of multinational
corporations in Spain," Strategic Management Journal, 11, (7), pp. 501-512.
Knoke, D. and JH. Kuklinski (1982), “Network anaylisis,” Sage, Newbury Park.
Miles, MB. and AM. Huberman (1994), “Qualitative data analysis: An expanded sourcebook,”
(2ª ed.). Sage, Thousand Oaks, CA.
Nohria, N. and S. Ghoshal (1997), “The differentiated network: Organizing multinational
corporations for value creation,” Jossey-Bass Publishers, San Francisco, CA.
Pentland, B. (1999), “Building process theory with narrative: From description to explanation,"
Academy of Management Review, 24, (4), pp. 711-724.
Prahalad, CK., and YL. Doz (1981), “An approach to strategic control in mncs," Sloan
Management Review, Summer, pp. 5-13.
Rowley, T., D. Behrens and D. Krackhardt (2000), “Redundant governance structures: An
analysis of structural and relational embeddedness in the steel and semiconductor
industries," Strategic Management Journal, 21, (3), pp. 369-386.
Scott, J. (1991), “Social network analysis: A handbook,” Sage, Newbury Park, California.
Silverman, D. (2001), “Interpreting qualitative data: Methods for analysing talk, text and
interaction,” (2ª ed.), Sage, London.
Subramaniam, M. and S. Watson (2006), “How interdependence affects subsidiary
performance," Journal of Business Research, 59, pp. 916-924.
Taggart, JH. (1999), “Subsidiary strategy: Concepts, empirical evidence,and policy implications,"
International Business Review, 8, (2), pp. 121-124.
Taggart, JH. and N. Hood (1999), “Determinants of autonomy in multinational corporation
subsidiaries," European Management Journal, 17, (2), pp. 226-236.
Teece, DJ., G. Pisano and A. Shuen (1997), “Dynamic capabilities and strategic management,"
Strategic Management Journal, 18, pp. 509-533.
Uzzi, B. (1996), “The sources and consequences of embeddedness for the economic performance
of organizations," American Sociological Review, 61, pp. 674-698.
Wasserman, S. and K. Faust (1999), “Social network analysis,” Cambridge University Press,
Cambridge.
Womack, JP., DT. Jones and D. Roos (1991), “The machine that changed the world: The story of
lean production,” Harper Collins Pub, New York.
IESE Business School-University of Navarra - 23
Yin, RK. (1994), “Case study research: Design and methods,” (2ª ed.), Sage, London.
Zukin, S. and P. DiMaggio (1990), “Introduction,” in P. DiMaggio (ed.), “Structures of capital:
The social organization of the economy,” Cambridge University, Cambridge, pp. 1-36.
24 - IESE Business School-University of Navarra